Cloud load balancing optimization method and system

A technology of load balancing and optimization methods, applied in transmission systems, digital transmission systems, electrical components, etc., can solve problems such as long network IO paths and low efficiency, achieve flexible flow control, eliminate locks and CPU context switching, and achieve high performance The effect of cloud load balancing

Active Publication Date: 2019-12-31
GUANGZHOU BINGO SOFTWARE
View PDF6 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In order to overcome the above-mentioned technical defects, the present invention provides an optimization method and system for cloud load balancing, which solves the long IO path of the traditional cloud load balancing network, as well as the low efficiency problems caused by data copying, lock overhead, CPU context switching, etc., and at the same time allows Cloud load balancing has the characteristics of flexible flow control, hot migration, elastic scaling, etc.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Cloud load balancing optimization method and system
  • Cloud load balancing optimization method and system
  • Cloud load balancing optimization method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] This embodiment also discloses an optimization method for cloud load balancing, including the steps of configuring cloud nodes, configuring virtual switches, and configuring cloud hosts. The cloud host is set in the cloud node and connected to the virtual switch.

[0059] Specifically, the configuration steps for cloud nodes are as follows: Figure 4 shown, including:

[0060] S101. Bind the network card of the cloud node to the VFIO driver.

[0061] With this setting, when the network card receives a packet, vfio shares the address of the network packet to the user space, and when the network card sends a packet, vfio shares the address of the user space network packet to the network card.

[0062] S102, the network card of the cloud node opens multiple groups of first sending and receiving queues through the DPDK process.

[0063] S103. The DPDK process of the cloud node opens a plurality of first polling thread PMDs, the first polling thread PMD is bound to the fi...

Embodiment 2

[0085] This embodiment discloses an optimization system for cloud load balancing, including a cloud node, a virtual switch and a cloud host: the cloud host is set in the cloud node and connected to the virtual switch.

[0086] Specifically, in this embodiment, the network card of the cloud node enables multiple groups of first sending and receiving queues through the DPDK process. The DPDK process of the cloud node opens multiple first polling thread PMDs, the first polling thread PMD is bound to the first sending and receiving queue one-to-one, the first polling thread PMD is bound to a CPU core, and the first polling thread PMD is bound to one CPU core. The PMD is used to acquire the data packets in the corresponding first sending and receiving queue in an active polling manner, and perform sending and receiving processing by the corresponding CPU core.

[0087]The virtual switch establishes a network bridge whose datapath type is netdev, and the user mode process correspond...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a cloud load balancing optimization method and system. Parameters in a cloud node, a virtual switch and a cloud host are configured respectively; the network card queue of thecloud node, polling threads of the cloud nodes, a DPDK port queue of the virtual switch, a vhost port queueof the virtual switch, network card queue of cloud host, and the load balancing service processes of the cloud host are in one-to-one correspondence, so that the data packet can be directly shared between the kernel layer of the cloud node and the load balancing service process of the cloud host; and meanwhile, the polling thread is bound with the cpu, and an active polling mechanism of the polling thread is adopted, so that the lock and the cpu context switch are eliminated. Compared with the prior art, the technical scheme provided by the invention has the advantages that the problems of long IO path of the traditional cloud load balancing network and low efficiency caused by data copying, lock overhead, CPU context switching and the like are solved, and meanwhile, the cloud load balancing has the characteristics of flexible flow control, thermal migration, elastic expansion andcontraction and the like. And high-performance cloud load balancing is realized.

Description

technical field [0001] The invention belongs to the technical field of load balancing, and in particular relates to an optimization method and system for cloud load balancing. Background technique [0002] The cloud load balancing service is a traffic distribution control service that distributes the access traffic to the backend multiple cloud hosts according to the forwarding policy in the cloud platform. It allows users to create cloud load balancing and configure the backend of the cloud load balancing service on the cloud platform. instance, and set relevant load balancing policies. Its principle diagram is as follows figure 1 shown. It has the characteristics of multi-tenancy, hot migration, and elastic scaling. [0003] Based on the above characteristics, the traditional implementation of cloud load balancing services is to run load balancing services in cloud hosts and provide services in a virtualized manner. Cloud Network provides features such as network acces...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): H04L29/08H04L12/24H04L12/803
CPCH04L41/0803H04L47/125H04L67/1001
Inventor 刘忻林冬艺肖兴
Owner GUANGZHOU BINGO SOFTWARE
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products